JP2000020710A - Image format converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve efficiency of image format conversion processing and to obtain an image with reduced quality deterioration. SOLUTION: The converter divides the picture data of one picture in a pixel block 13 to be interpolated into prescribed pixel blocks and obtains a pixel block 14 obtained by shifting one pixel in the Y direction to execute linear interpolation by adjacent pixel data. Namely a pixel block 15 is obtained by executing interpolation processing between both pixel blocks 13, 14. Then pixel data arrayed on the lowermost stage of the pixel block 15 are copied.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image format conversion apparatus capable of improving the efficiency of image format conversion processing and obtaining an image with little deterioration in image quality.

[0002]

2. Description of the Related Art In recent years, a method of dividing image data of one screen into predetermined pixel blocks and performing image processing for each pixel block has been used as an image processing method. In particular, when used for image communication and the like, there is a demand for reducing the amount of image data and efficiently processing more data in a short time. Therefore, after performing image processing on an image in which the number of pixels in the vertical and horizontal directions is 1 / N, the image is copied in the main scanning X direction and the sub-scanning Y direction, enlarged to N times, and restored to the original image size. Have been proposed.

FIG. 11 is a schematic diagram of a four-fold enlargement process according to the conventional method 1, which is performed when the image size is quadrupled by the image format conversion process.
This enlarges the × 8 pixel block 1 to the 16 × 16 pixel block 2. By copying one pixel each in the main scanning X direction and the sub scanning Y direction, the image is enlarged four times. Explaining the details of the enlargement processing, the pixel 1 (x1, y1) of the 8 × 8 pixel block 1 is (x1, y1), (x1, y2), (x2, y) of the 16 × 16 pixel block 2.
1) Copy to the positions of (x2, y2). Similarly, the format of the other pixels is quadrupled by copying them to the lower, right, and lower right positions.

In another conventional method 2, when image data of one screen is divided into predetermined pixel blocks and image processing is performed, all of the image processing for each pixel block and the output processing for each pixel block have been completed. Later, there was a method of performing image format conversion processing and one-screen filter processing for removing noise on image data of one screen.

FIG. 12 is a flowchart of an image format conversion process according to the conventional method 2. The pixel block input to the pixel block input unit 3 for each pixel block is processed by the pixel block processing unit 4 and output to the pixel block output unit 5. When the pixel block end determination unit 6 determines whether the pixel block output processing has been completed for one screen, the image format conversion processing is performed on the image data of one screen by the format conversion unit 7, and then the one screen filter processing unit 8 Performs the filtering process.

[0006]

In the above-mentioned conventional system 1, when the image is simply enlarged by N times, the image quality is deteriorated, in particular, images and lines in oblique directions appear to be stepped.
In addition, when only linear interpolation is simply performed, it is necessary to store pixel data for one scan in order to interpolate pixel data at the lowest stage of a pixel block.

Further, in the conventional method 2, since the image format conversion processing and the one-screen filter processing are performed on the image data of the entire one screen, the memory area to be used increases, and the data once output to the outside is again stored inside. There is a problem that processing time is required because the filter processing is performed by taking in the data.

An object of the present invention is to solve the above-mentioned problems, and to provide an image format conversion apparatus capable of improving the efficiency of image format conversion processing and obtaining an image with little deterioration in image quality.

[0009]

According to an aspect of the present invention, there is provided an image format conversion apparatus which divides image data of one screen into predetermined pixel blocks, and converts the image data into a predetermined pixel block. The pixel block interpolation processing is performed by linearly interpolating and copying the pixel data at the bottom of the pixel block. According to the present invention, it is possible to obtain an image with little deterioration in image quality.

An image format conversion apparatus according to a second aspect of the present invention divides image data of one screen into predetermined pixel blocks, and is provided at a subsequent stage of a pixel block processing unit for processing an image for each pixel block. Thus, pixel block interpolation processing, image format conversion, and pixel block data output are performed for each pixel block. According to the present invention, there is an effect that an image format conversion process can be efficiently performed while suppressing an increase in a used memory area, and an image with less image degradation can be obtained.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

(Embodiment 1) FIG. 1 is a diagram showing an image format conversion process of an image format conversion apparatus according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 9 denotes one pixel block when image data of one screen is divided into 8 × 8 pixel blocks, and 10 denotes 16 × 1 when the image data is quadrupled.
It is a 6 pixel block. When the image format conversion processing device according to the first embodiment is used, the 16 × 1
In the 6-pixel block, pixel data from pixel 1 (x1, y1) to pixel 64x (x16, y15) is composed of original pixel data and pixel data obtained by performing linear interpolation, and the pixel at the bottom of the pixel block Data (pixel 57
(X1, y16) to pixel 64x (x16, y16)) are configured by copying the upper pixel data (from pixel 57 (x1, y15) to pixel 64x (x16, y15)).

Next, details of a case in which one frame of a QCIF image is divided into 8 × 8 pixel blocks, image-processed, and enlarged to a CIF image using the image format conversion processing apparatus of the present invention will be described with reference to FIGS. This will be described with reference to FIG.

FIG. 2 is an output diagram of the original pixel block data. First, the pixel data of the 8 × 8 pixel block 11 is 1
Main scan X direction, sub scan Y to 6 × 16 pixel block 12
The pixels are arranged at intervals of one pixel in each direction.

Next, the interpolation processing in the sub-scanning direction will be described with reference to FIG.
A description will be given with reference to FIG. First, the pixel data (from pixel 9 (x1, y2) to pixel 64 (x8, y8)) of the second to eighth rows (y2 to y8) of the pixel block 13 to be interpolated in FIG. , Which is the bottom row of the same pixel block 13
The pixel block 14 of 8 × 8 pixels is created by combining the pixel data of the row (y8) (from the pixel 57 (x1, y8) to the pixel 64 (x8, y8)). Then, linear interpolation is performed on the pixel data at the same position in the original pixel block 13 and the newly created pixel block 14, and the sub-scan Y
A pixel block 15 interpolated in the direction is created. This means that linear interpolation has been performed on the original pixel data and the data shifted by one pixel in the sub-scanning Y direction.

However, in this case, taking advantage of the fact that the human eye is less sensitive to the image in the sub-scanning direction than in the main scanning direction, the maximum of each pixel block can be saved without storing the pixel data for one scan. The pixel data in the lower part (y8) is copied and arranged.

FIG. 4 shows an output diagram of pixel block data after interpolation in the sub-scanning direction.
5 is arranged in a 16 × 16 pixel block 16. The arrangement at this time is arranged at a position shifted by one pixel in the sub-scanning Y direction from the position arranged in the 16 × 16 pixel block 12 of FIG. As a result, the pixel data interpolated between the pixels adjacent in the sub-scanning Y direction is embedded between the two original pixel data.

Here, the pixel block to be processed is one of the screens.
If the pixel block is not the leftmost pixel block, an interpolation process in the main scanning X direction is performed. FIG. 5 shows a pixel block interpolation processing diagram in the main scanning X direction. 5 is a QCIF image of one screen, 1 in FIG.
Numeral 8 denotes two pixel blocks adjacent to each other in the main scanning X direction. An extracted 8 × 8 pixel block on the left side of the pixel block 18 is an interpolation target pixel block 19, a left pixel block and a right pixel block. A pixel block 20 is obtained by extracting an 8 × 8 pixel block from the interpolation target pixel block 19 by shifting one pixel in the main scanning X direction using the above. Then, the original pixel block 19
Then, linear interpolation is performed using the pixel data at the same positions of the newly created pixel block 20 and the pixel block 21 that has been subjected to the interpolation processing in the main scanning X direction. This allows
This means that linear interpolation has been performed on the original pixel data and the pixel data shifted by one pixel in the main scanning X direction.

FIG. 6 shows an output diagram of pixel block data after interpolation in the main scanning X direction. The 8 × 8 pixel block 21 after interpolation is arranged in the 16 × 16 pixel block 22. The arrangement at this time is arranged at a position shifted by one pixel in the main scanning X direction from the position arranged in the 16 × 16 pixel block 12 in FIG. As a result, the pixel data interpolated between pixels adjacent in the main scanning X direction is embedded between the two original pixel data.

Thereafter, the pixel block 21 (see FIGS. 5 and 6) obtained by interpolation in the main scanning direction is subjected to interpolation processing in the sub-scanning Y direction (see FIG. 3). Thus, linear interpolation is performed between pixel data obtained by interpolating original pixel data in the main scanning X direction and data obtained by further shifting the pixel data after the main scanning X direction by one pixel in the sub scanning Y direction. become.

FIG. 7 shows an output diagram of pixel block data after interpolation in the main scanning X direction and sub scanning Y direction.
The eight pixel block 23 is arranged in the 16 × 16 pixel block 24. The arrangement at this time is arranged at a position shifted by one pixel in the sub-scanning Y direction from the position arranged in the 16 × 16 pixel block 22 in FIG. As a result, the pixel data interpolated in the main scanning X direction is further interpolated in the sub-scanning Y direction and embedded between the two pixel data, and the image format conversion to the CIF size of the pixel block is completed.

Next, FIG. 8 shows a pixel block interpolation process in the main scanning X direction of the right end block. When the pixel block to be processed is the rightmost pixel block on the screen, the same as the sub scanning Y direction interpolation process is performed. The right side of the pixel data (from pixel 2 (x2, y1) to pixel 64 (x8 to y8)) of the second to eighth columns (x2 to x8) of the rightmost interpolation target pixel block 25 is the same as The pixel data (pixel 8 (x8, y) of the eighth column (x8) which is the rightmost column of the pixel block 25
1) to 64 × (x8, y8)
A pixel block 26 of eight pixels is created. Then, linear interpolation is performed on the pixel data at the same position of the original pixel block 25 and the newly created pixel block 26 to create a pixel block 27 that has been interpolated in the main scanning X direction. This means that linear interpolation has been performed between the original pixel data and the pixel data shifted by one pixel in the main scanning X direction. However, in this case, since the pixel block is the rightmost pixel block on one screen, there is no pixel block adjacent in the main scanning X direction. Therefore, the pixel data in the rightmost column (x8) of the pixel block is copied and arranged as it is. And Thereafter, the interpolated 8 × 8 pixel block 21 (see FIG. 6) is converted into a 16 × 16 pixel block.
It is arranged in the pixel block 22 (see FIG. 6).

Subsequently, in the sub-scanning Y direction, an interpolation process is performed in the same manner as in the pixel block other than the right end of the screen (FIG. 3).
7), and arranged at the position indicated by the 16 × 16 pixel block 24 in FIG.

When this processing is repeated and the interpolation processing is completed for all the pixel blocks on one screen, the QCIF
The image format conversion from the image to the CIF image ends.

(Embodiment 2) FIG. 9 is a schematic diagram of an image format conversion processing apparatus according to Embodiment 2 of the present invention. In FIG. 9, reference numeral 28 denotes a pixel block input unit;
Denotes a pixel block processing unit, 30a, 30b, 30c, 30
d is a pixel block output unit, 31a and 31b are sub-scanning Y-direction interpolation processing units, 32 is a main scanning X-direction interpolation processing unit, and 33 is a pixel block end determination unit.

Next, the operation will be described. The pixel block input section 28 and the pixel block processing section 29 perform the processing for each pixel block as in the prior art, and then perform the processing for the pixel block output section 30.
a pixel block output and sub-scanning Y-direction interpolation processing unit 3
The interpolation processing of 1a is performed. Hereinafter, the pixel block output unit 30
b pixel block output and main scanning X direction interpolation processing unit 3
(2) Image format conversion by sequentially processing the pixel block output of the pixel block output unit 30c, the pixel block output of the sub-scan Y-direction interpolation processing unit 31b, and the pixel block output of the pixel block output unit 30d in pixel block units. I do. That is, the pixel block end determination unit 3
When it is determined that all the pixel blocks have been completed according to 3, it means that image output for one screen and image format conversion with interpolation processing have been completed.

Next, details of a case in which one frame of a QCIF image is divided into 8 × 8 pixel blocks, image-processed, and enlarged to a CIF image using the image format conversion processing apparatus of the second embodiment are shown in FIG. This will be described with reference to FIGS. Here, description will be made assuming that the method of the block interpolation processing uses the method of the first embodiment of the present invention.

FIG. 10 shows a flowchart of the image format conversion processing according to the second embodiment. In FIG. 10, 34 is a pixel block input unit, 35 is a pixel block processing unit, 36a and 36b are pixel block output units,
36c and 36d are previous pixel block output units, and 36e and 36
f is a current pixel block output unit, 37 is a sub-scan Y direction interpolation processing unit, 38 is a pixel block position determination unit, and 39 is a main scan X
A direction interpolation processing unit, 40 is a pixel block end determination unit, 41
Denotes a pixel block data storage unit.

First, an external QCIF image is fetched into the pixel block input unit 34 by 8 × 8 pixel blocks, and the pixel block processing unit 35 performs pixel block processing. Then, the pixel data of the 8 × 8 pixel block 11 (see FIG. 2) is converted into a 16 × 16 pixel block 12 (see FIG. 2) by one pixel each in the main scanning X direction and the sub-scanning Y direction. Output to the output unit 36a.

Next, the operation of the interpolation unit 37a in the sub-scanning Y direction will be described with reference to FIG. First, the second to eighth rows (y
Below the pixel data (pixels 9 (x1, y2) to pixel 64 (x8, y8)) of the second row (y8), the pixel data of the eighth row (y8) which is the lowermost row of the same pixel block 13 57 (x1, y8) to pixel 64 (x8, y8)
To create a pixel block 14 of 8 × 8 pixels. Then, linear interpolation is performed on the pixel data at the same position of the original pixel block 13 and the newly created pixel block 14 to create a pixel block 15 that has been subjected to interpolation processing in the sub-scanning Y direction. As a result, the original pixel data and the data shifted by one pixel in the sub-scanning Y direction are subjected to linear interpolation, and are output to the pixel block output unit 36b.

However, in this case, utilizing the fact that the human eye is less sensitive to the image in the sub-scanning direction than in the main scanning direction, the pixel data of each pixel block can be stored without storing the pixel data for one scan. The pixel data at the bottom (y8) is copied and arranged.

Thereafter, the interpolated 8 × 8 pixel block 15 shown in FIG. 4 is arranged in a 16 × 16 pixel block 16.
The output at this time is the 16 × 16 pixel block 12 shown in FIG.
Is output at a position shifted by one pixel in the sub-scanning Y direction from the position arranged at. As a result, the pixel data interpolated between the pixels adjacent in the sub-scanning Y direction is embedded between the two original pixel data.

Here, the pixel block position determination unit 38a determines whether the pixel block currently being processed is the leftmost pixel block on the screen. If the pixel block is the leftmost pixel block, the pixel block data is stored in the pixel block data storage. Part 41
The input of the next pixel block data and the processing of the pixel block are performed by the pixel block input unit 34 and the pixel block processing unit 35.

If the pixel block is not the leftmost pixel block, the interpolation processing section 39a in the main scanning X direction performs an interpolation process. Main scan X
Details of the direction interpolation processing will be described with reference to FIG. In FIG. 5, reference numeral 17 denotes a QCIF image of one screen, and reference numeral 18 denotes two pixel blocks adjacent in the main scanning X direction. An 8 × 8 pixel block extracted from the left 8 × 8 pixel block of the pixel block 18 is shifted by one pixel in the main scanning X direction using the pixel block 19 and the left and right pixel blocks. Is a pixel block 20. Then, linear interpolation is performed using pixel data at the same position in the original pixel block 19 and the newly created pixel block 20, respectively, to create a pixel block 21 that has been subjected to interpolation processing in the main scanning X direction. This means that linear interpolation has been performed between the original pixel data and the pixel data shifted by one pixel in the main scanning X direction. After that, 8 ×
The 8-pixel block 21 (see FIG. 6) is output to the 16 × 16 pixel block 22 (see FIG. 6). The output of the previous pixel block output unit 36c at this time is output to a position shifted by one pixel in the main scanning X direction from the position output to the 16 × 16 pixel block 12 in FIG. As a result, the pixel data interpolated between pixels adjacent in the main scanning X direction is embedded between the two original pixel data.

Thereafter, the sub-scanning Y-direction interpolation processing section 37b performs an interpolating process on the pixel block 21 (see FIG. 5) obtained by interpolating in the main scanning X-direction (FIG. 3).
reference). Thus, linear interpolation is performed between pixel data obtained by interpolating original pixel data in the main scanning X direction and data obtained by further shifting the pixel data after the main scanning X direction by one pixel in the sub scanning Y direction. become. After that, 8
The previous pixel block output unit 36d converts the × 8 pixel block 23 (see FIG. 7) into the 16 × 16 pixel block 24 (see FIG. 7).
Output to The output at this time is output to a position shifted by one pixel in the sub-scanning Y direction from the position arranged in the 16 × 16 pixel block 22 in FIG. As a result, the de-pixel data interpolated in the main scanning X direction is further interpolated in the sub-scanning Y direction and embedded between the two pixel data, and the image format conversion of the previous pixel block to the CIF size is completed. .

After the output of the pixel data obtained by interpolating the previous pixel block is completed, the pixel block position determining unit 38b determines whether or not the current pixel block is the rightmost pixel block on the screen. If it is not a pixel block, the process proceeds to the pixel block input unit 34 for the next pixel block input processing. In the case of the right end block, the pixel data (pixel 2 (x2, y1)) of the second to eighth columns (x2 to x8) of the right end pixel block 25 (see FIG. 8) are similar to the interpolation processing in the sub-scanning Y direction. To pixel 64 (x8 to y8)
The pixel data of the 8th column (x8), which is the rightmost column of the same pixel block 25 (from pixel 8 (x8, y1) to pixel 64 (x8, y8)), is a pixel of 8 × 8 pixels Block 26 (see FIG. 8) is created. Then, a linear interpolation is performed on the pixel data at the same position of the original pixel block 25 and the newly created pixel block 26, and the pixel block 27 which has been subjected to the interpolation processing in the main scanning X direction by the main scanning X direction interpolation processing unit 39b. (See FIG. 8). This means that linear interpolation has been performed between the original pixel data and the pixel data shifted by one pixel in the main scanning X direction. However, in this case, since it is the rightmost pixel block of one screen, there is no pixel block adjacent in the main scanning X direction.
Pixel data of the rightmost column (x8) of the pixel block is copied as it is and output to the current pixel block output unit 36e. Thereafter, the interpolated 8 × 8 pixel block 21 (see FIG. 6) is output to the 16 × 16 pixel block 22 (see FIG. 6).

Subsequently, in the sub-scanning Y-direction, an interpolation process is performed by the sub-scanning Y-direction interpolation processing unit 37c in the same manner as in the pixel block not at the right end of the screen (see FIG. 3).
Output to the current pixel block output unit 36f at the position indicated by the × 16 pixel block 24.

This process is repeated, and if it is determined by the pixel block end determining unit 40 whether the interpolation process has been completed for all the pixel blocks on one screen, the CI from the QCIF image is determined.
The image format conversion to the F image ends.

[0039]

As described above, according to the present invention, it is possible to prevent the deterioration of the image quality, in particular, that the image or the line in the oblique direction looks like a staircase, the increase of the memory area used, and the efficiency of the image format conversion processing. As a result, the processing time can be reduced. Therefore, a great effect is brought about in increasing the amount of data (the number of frames) that can be processed within a predetermined time.

[Brief description of the drawings]

FIG. 1 is an image format conversion processing diagram of an image format conversion processing device according to a first embodiment of the present invention.

FIG. 2 is an output diagram of original pixel block data.

FIG. 3 is a diagram showing a pixel block interpolation process in the sub-scanning Y direction.

FIG. 4 is an output diagram of pixel block data after interpolation in the sub-scanning Y direction;

FIG. 5 is a diagram showing a pixel block interpolation process in the main scanning X direction.

FIG. 6 is an output diagram of pixel block data after interpolation in the main scanning X direction.

FIG. 7 is an output diagram of pixel block data after interpolation in a main scanning X direction and a sub scanning Y direction.

FIG. 8 is a diagram showing a pixel block interpolation process in the main scanning X direction of the right end block.

FIG. 9 is a schematic diagram of an image format conversion processing device according to a second embodiment of the present invention.

FIG. 10 is a flowchart of an image format conversion process according to the second embodiment of the present invention.

FIG. 11 is a schematic diagram of a 4 × enlargement process according to the conventional method 1.

FIG. 12 is a flowchart of an image format conversion process according to the conventional method 2;

[Explanation of symbols]

11, 15, 21, 23 8 × 8 pixel block 12, 16, 22, 24 16 × 16 pixel block 13 Pixel block for interpolation processing 14 13 Pixel block 15 13 and 14 shifted by one pixel in the sub-scanning Y direction Interpolated pixel block 17 QCIF image of one screen 18 Pixel block adjacent in main scanning X direction 19 Pixel block to be interpolated (previous pixel block) 20 Pixel block 21 19 and 20 shifted by one pixel in main scanning X direction from 19 25, a pixel block shifted by one pixel in the main scanning X direction from pixel 25 27, a pixel block interpolated by 25 and 26, 28, 34 pixel block input unit 29, 35 pixel block Processing unit 30, 36 Pixel block output unit 31, 37 Sub-scan Y-direction interpolation processing unit 32, 39 main scanning X direction interpolation processing unit 33, 40 pixel block end determination unit 38 pixel block position determination unit 41 pixel block data storage unit

Claims (2)

[Claims] 1. A pixel block interpolation process is performed by dividing image data of one screen into predetermined pixel blocks, performing linear interpolation with adjacent pixel data, and copying the pixel data at the bottom of the pixel blocks. An image format conversion apparatus characterized by performing the above. 2. A pixel block interpolating process, an image format conversion, and a pixel block data are provided at a subsequent stage of a pixel block processing unit that divides image data of one screen into predetermined pixel blocks and processes an image for each of the pixel blocks. Outputting the image data for each of the pixel blocks.